okay — so, I pooh-poohed the so called issue with G+ sharing but have now run into this with our CaliforniaNewswire.com site after moving from woo canvas to newsmag 1.6 this past weekend.
Previously and the past number of years if I shared a story on G+ it would grab photo, headline, and first paragraph … now with newsmag it only grabs photo and headline … super annoying.
In trying to see “what up with dat” … the only notable diff I could see “on page” was the “categories” being above headline, whereas on woo, they follow the author byline and date … as in
Story by Name Here on XX day, posted in [cat] [cat] [cat]
I only bring this up as I will be testing MOVING the “categories” *BELOW* the byline for a story and/or turning them off, to see how that impacts the G+ share.
So, those of you who said something up …. I take a step back and no longer call you crazy 😉
Hi,
Thanks for letting us know bout this. I have tested this on our demo and found the same behavior: http://screencast.com/t/wu7zYIWJSh
I will add this on our todo list for further testing it and try to find a solution for it.
Let us know if you have any update on this.
Thanks
Hi Chris,
I’ve did some research on this and found that by default, google+ removed the description from their sharer. Now you need to have interactive posts with call to action buttons for this.
You can find more details about google interactive posts here: https://developers.google.com/+/web/share/interactive ..and use a plugin to create your interactive posts: http://searchwilderness.com/new-google-plus-interactive/
Thanks
Actually … that is separate function for a different kind of content sharing.
Here is example of what I’m talking about …
if you go to your G+ account and post an update, and paste in this link from one of our sites — one of the last ones running Woo Canvas vs Newsmag:
http://newyorknetwire.com/2015/04/20/newswire_NYNW150338-9452.php/new-easeus-partition-master-10-5-able-convert-mbr-disk-gpt-disk-convert-gpt-disk-mbr/
this will populate the photo, headline and first paragraph in the content box.
If I do that on CaliforniaNewswire.com with a story, running Newsmag 1.6, it no longer does that; although it did it a week ago prior to the switch to Newsmag from our old Woo custom template.
This leads me to believe all things being equal that there is some element of the rendered page structure which causes Google’s scraper to not pull the full text in quite the same way.
——–so———–
I did try removing the “categories” from above headline/title on CaliforniaNewswire.com and this had no effect. I am going to try some other things like “simple sidebar” and removing related content, social media buttons, etc., to see if it’s “relational” content confusing it (e.g., meta data or mark-up related to the other images/titles on page), or what. Super annoying.
It may be that having the bulk of other headlines/titles/images inside the “woo sidebar” separates the onpage content enough to make the Google scraper less confused.
They really should be pulling the damn meta/og description tag, and no good reason for them to still be “scraping” the page.
Could be something as simple as they don’t like the “full width headline” which is in separate “block” from the body text; meaning not in the same “column” …. (sigh)
I’m playing with it, as time permits.
Upshot — the Woo Canvas template we’ve had for several years DOES WORK for G+ capture, while the Newsmag layout I’ve setup does NOT … so, I’m always frustrated by these micro-management issues with Google …
And don’t get me started on the right format for sitelinks for CSE hinting. Lack of documentation on that is ridiculous from them. 🙂
Anecdotal…
tested a Rolling Stone article with wide headline
http://www.rollingstone.com/movies/features/secrets-of-avengers-age-of-ultron-joss-whedon-tells-all-20150422
and it also skips the first paragraph, but grabs the sub-headline which is the same “block” (basically two cols vs one col). So it might be the wide headline template alone causing that oddity.
I am still playing with this 🙂
Finally tagDiv know there’s a problem.
Many users had reported before but no one think it is the theme’s problem.
Just to clarify, it’s not really a ‘theme’ issue – for those curious. It’s an issue with Google’s bot having issues with scraping pages to verify actual content vs spammers who try to trick google with content in META data that doesn’t match what is actually on the page.
So, this is why it “seems” like Google’s bot is just ignoring the “description” tags; they use that as “hint” then read the page and look for stuff in certain common “blocks” of content, like “first full paragraph of text on page” …
I am trying to sort out why this happens on some templates, layouts and not on others, across various top websites and small ones, too.
For example I tested page from WIRED mag the other day, and it grabbed the first paragraph; where I tested a page from our main news service portal — which does not use WordPress or any popular CMS [in house system we built in 2005 and I built the templates myself from scratch using customized bootstrap 2.3x code] — and these “should” work but also don’t get the first paragraph captured by G+.
(e.g. this page “should” work — I built it myself 🙂 but does not:
https://www.send2press.com/newswire/tickets-on-sale-for-music-festival-drawing-big-name-acts-to-ohio-aiding-cancer-charities-2015-0416-03.shtml )
this page on wired.com does NOT work (odd they have no G+ buttons; bad SEO!)
http://www.wired.com/2015/04/porsches-new-boxster-made-just-driving-enthusiasts/
So far does NOT appear to be these specific elements, with NewsMag
a) “wide screen” headline — across top, 2 columns
b) category buttons/links above headline, after breadcrumbs (on/off no diff)
c) not issue with using mod_pagespeed for comment and whitespace removal, or rewriting images
d) not issue with https vs non https
*anecodatal: have run into couple of issues with Google News, when scraping page couple of times they grabbed the category buttons in place of the story headline or appended them in front of headline — contrary to news XML sitemap — stupid, stupid, stupid! Possibly need more “separation” from those and headline — not yet sure on that.
So, I’m playing with it in the midst of doing major security audit of WordPress plugins, and code across 15 websites (ugh). Only time we had a WP website “hacked” was way back (something like v 1.6 or 2.6 or something? — Al Gore’s website hacked same day ours was …Nov 2007 for those who know their beans). So, we stay on top of security stuff pretty strictly but these new vulnerabilities are a nest of vipers waiting to get milked for their toothy sting juice (have fun with that sentence English translators!). 😉
Hi,
Thanks Chris for all this feedback and updates on this. I’ve tested this myself and found that on some older posts the description is actually pulled by google and it works fine using Yoast’s og:description meta, however on some other posts it doesn’t even though the description tag is present. This is strange as the seems to have the same open graph tags and the same structure the g+ sharer doesn’t add the description.
We’ll dig a little deeper into this and see if we find a reason or a fix on this in next theme versions.
Thanks
Funnily enough Google’s own G+ buttons on their Google Alert emails for news, clearly showing the headline, photo and first paragraph, won’t even grab first paragraph consistently.
I’m almost wondering if it’s placement of an “article” tag, not using an article tag at all, or applying an “ID” of “entry” to the entire post block similar to ID=”header” to the head area.
Playing with it more … we’ll be kings if we solve this ! But not going to stay up nights worrying about it.
I wish it can be solved soon.
It still happen in Newsmag 2.2
Hi …
fyi, G+ having all sorts of sharing issues unrelated to any specific theme, website, or content.
I have noticed this month issues in sharing from Google Alerts, sharing directly from both WP and non-WP sites, — and from about a dozen very high-profile top 1000 web properties websites, where in some cases only a photo is captured and NO text data — and this is on both WP and non-WP sites.
For example today, sharing a story on our main site from Google alert would only grab photo; however sharing it from Google News captured photo and the headline, date, etc. (but not first paragraph).
I have seen almost *no* website able to populate the pargraph/excerpt box lately, so there is some fundamental issue with Google that is not theme or site or content specific.
Even the example I’d posited in this thread initially — using Canvas on one of our legacy sites (moving to Newsmag for 2016!) — also right will only grab photo and no other data.
SO — upshot, not sure there is any inherent “fix” for this issue as it seems Google has some oddity in this with their system which comes, goes, or will never be fixed.
It’s one of those things I think folks need to not stress over too greatly, as Google doesn’t seem to know what to do with G+ at this point.
-
This reply was modified 10 years by
simchris.
